1
FDA Oncology Drug Approval
Endpoints, Effectiveness, and ApprovalGerald H. Sokol MD, MS, FCP* and Robert Kane, MD, FACP
Division of Drug Oncology ProductsOffice of Oncology Drug Products
CDER, FDANot an official FDA policy
Asst. Prof Medicine and Clinical PharmacologyUniformed Services University or the Health Sciences
2
The medical reviewer in his natural habitat
3
4
The “White Oak” Consolidation
5
6
FDA - Oncology Drugs
• Not – Drug imports from Canada
• Not – Vioxx, Celebrex, Plan B
• Not – dietary supplements -Foods
• No stock market tips
7
FDA Oncology Drug Approval
• FDA – U.S. Food and Drug Administration– In the Dept of HHS – Executive Branch
• Created by Congress because of prior unsafe drugs being marketed
• FDA charged by Congress to evaluate all prescription drugs seeking marketing in the U.S.
• Federal Laws, Regs govern these activities
8
Communication is the Key
9
Overview for oncology drug approval
• Evidence for efficacy and safety• Good Evidence = Approval • Endpoints for trial design• Endpoints for FDA approval• Trial design issues for efficacy• Trial results analysis• “Targeted” therapy approval and problems
Interplay of disease state, existing Rx options, endpoint options, strength of evidence
10
Regulatory Perspective on Drug Development and Approval
Pre IND
1 2 3 4
Non – clinical studies
~~~~FDA Consultation~~~~~~~~~~~~~~~~~~~
IND Filing NDA Filing
Comprehensive multidisciplinary review often with
Advisory Committee discussion
Monitor safety, review new protocols, annual
reports, approve exceptions
Safety and Phase 4 monitoring
Regular Accelerated
NDA Filing
Approval options
11
Drug Development Focal Points -FDA Meetings
Phase Format Intent (FDA concerns)
Pre-IND / IND
T-con, FTF, none
Ph 1 design – FDA safety -population and dosing
EOP1 Tcon, FTF Ph 2 design – FDA safety - population and dosing
EOP2 FTF Ph 3 design – FDA safety, study design & analysis
Pre-NDA FTF
(face to face)
Results; format & content of reports; time frame of submission; priority?
12
FDA - Oncology Drugs
• Office of Oncology Drug Products – OODP• Three divisions
DDOP- Chemotherapy drugs for Cancer
DBOP- Biologic oncology therapies BLAs
DMIHP- Medical Imaging and hematology
13
Requirements for Drug Approval
U.S. Statutes –Congress
Labeling 1906 Pure Food & Drug Act
Safety 1938 Food, Drug, and Cosmetic Act
Efficacy 1962 FDC Amendments Harris-Kefauver
FDAMA Amdnts. 1997
Code of Federal Regulations (CFRs)
Regulations from FDA provide interpretation of the laws
14
Drug Approval Tracks
• FDA Modernization Act 1997 (FDAMA): CFR 312 subpart E regulations
• To expedite development for life-threatening illnesses
– Fast track • process for meeting FDA -unmet med need
– Priority review • Determined by FDA after NDA submitted• 6 month NDA review time frame for products
addressing unmet medical need
15
Drug Approval Tracks (2)
• Accelerated approval “CFR 314 subpart H” – 1992
• Special Protocol Assessment – agreement between FDA and sponsor on
clinical protocols for phase 3 studies forming primary basis for demonstrating efficacy for NDA; 45 day clock
16
FDA Oncology Drug Approval
• How do you achieve approval ?
• Provide substantial evidence of
Efficacy and Safety
• What is substantial evidence for Effectiveness– “A & WC investigations” CFR 314.126
• 2 or more studies
– required by 1962 amend. to FFDCA Source = Controlled clinical trials
17
Basis for New Drug Approval
• Demonstration of efficacy with acceptable safety in adequate and well-controlled studies CFR 314 - NDA Regulations
• Ability to generate product labeling that– Defines an appropriate patient population for
treatment with the drug– Provides adequate information to enable safe
and effective use – prescribing of the drug
• Analogous rules for Biologics - BLA
18
FDA Oncology Drug Approval
• Clinical trials:– When we do not know which therapy is better– Patients are fully informed of the uncertainty
about which therapy is better and give consent– We can compare one treatment with another in
a controlled way– Today’s “standard” therapy was last year’s
investigational treatment
19
FDA Oncology Drug Approval
• Controlled clinical trials:– Can be verified, repeated if necessary– Can allow us to be convinced a new therapy
is effective – or it is not.– Allow us to compare how well tolerated -
(how safe) a new therapy is versus a standard
• Efficacy with Safety => Approval
20
Some trial considerations
• Usual situation for many FDA drug approvals– Multiple studies– Studies are large 1,000 – 5,000 patients– Placebo control group– Double Blinded or a blinded independent assessment– Highly significant p values (0.001)
• Oncology drug data submitted for approval– One study, 100 – 800 patients– No blinding, no placebo control– Heterogeneous patient group– Statistical evidence ~ 0.03 – 0.05*Concern – how confident can we be that results are real
21
The FDA mission is not a new idea!
"The aim ... is not simply to accept the statements of others, but to investigate the causes that are at work in nature."
Albertus Magnus, de Mineralibuscirca 1250
22
FDA Oncology Drug Approval
• Suppose YOU are the FDA –
• What benefits should a new drug have to allow its marketing approval in the U.S.?
• How “safe” should the drug be?
23
FDA Oncology Drug Approval
• What benefits should a new drug have for marketing approval in the U.S.?
• LIVE LONGER ---------------------Effective
• LIVE BETTER – QUALITY ------Effective
24
FDA Oncology Drug Approval
• What benefits should a new drug have for marketing approval in the U.S.?
• LIVE LONGER ---------------------Effective• LIVE BETTER – QUALITY ------Effective• Improved safety - with efficacy• Benefits outweigh Risks (312.84)
• B / R assessment - in disease context
• COST ? Less expensive? / Reimbursement ?– Not purview of FDA
25
FDA Oncology Drug Approval
• Judgment of Benefits versus Risks
• But - NEVER Have ALL the Data
• Some benefits or adverse effects occur:– Rarely– After long time interval
• How long to study and wait before Approval
• Too slow or too fast to approve?
26
What Drugs Are Safe ?
27
What Drugs Are Safe ?
None of them
28
What Drugs Might Help Someone?
29
What Drugs Might Help Someone?
All of them
30
Can we tell in advance who might be helped and
who might be hurt by a drug?
31
Can we tell in advance who might be helped and
who might be hurt by a drug?
NO
But we hope to be able to soon
32
Two Types of Drug Approval: Regular or Accelerated
Endpoints Supporting Regular Approval Regular Approval Demonstrate Clinical Benefit– Longer life – Better life (relief of tumor-related Sx) - PRO
• Requires a valid measure of how a patient feels or functions
– Favorable effect on established surrogate
33
Accelerated Approval (AA)
Only applies in the setting of a new drug for a serious or life-threatening illness:
• Improvement over available therapy
• Study may use a surrogate endpoint, reasonably likely to predict clinical benefit
• Requires confirmation of benefit
Fed Register 1992
34
Oncology Trial Endpoints (1)
CLINICAL BENEFIT ENDPOINTS• LIVE LONGER – Measure Survival- OS
Efficacy, reassures for safety, unbiased endpt but – subsequent therapy, long time required
• LIVE BETTER – Measure QOL – PROImportant: control group + blinding needed (bias)
hard to measure, scale problems, missing data
Drug Tox. symptoms versus tumor symptoms
=> COMPARISONS ARE NECESSARY
Note – These endpoints do not measure the tumor
35
Tumor measurement endpoints
Randomize Response Progression Death
or recurrence
I__________I__________I__________ITime to Response Response Duration
Metastatic - Time to Progression (TTP) ---- PFS
Adjuvant - Time to Recurrence ---- DFS
Overall Survival
36
Oncology Endpoints (2)
• DISEASE-FREE SURVIVAL (DFS)
– Composite of survival and NED – Recurrence is associated with symptoms,
new therapies, cognitive effects
– Adjuvant treatment setting
• Breast cancer
• Colorectal cancer–3 year DFS (p ≤ 0.03) => 5 year OS
• New areas likely – lung, prostate, brain, etc
37
Endpoints - Regular Approval (3)
• Survival
• QOL – PRO• Safety improved- with efficacy
• DFS – adjuvant and leukemia settings
• Improvements convey/are clinical benefits
• Improvements show effectiveness
Demonstrate these endpoints
= (Full) Regular Approval (RA)
38
Tumor assessment endpoints (5)
• Response - Response rate (RR) = Drug activity (not the same as efficacy)
anatomic imaging – measure tumor - RECIST criteriaComplete responses (CR) – stronger evidence
• Time to progression (TTP) Progression – anatomic imaging - RECIST based Includes stable disease (natural history)
• Progression-free survival (PFS) – Composite of progression (TTP) and Death– Includes stable disease [ODAC (preliminary) – several months’ difference]
39
Tumor assessment endpoints (6)
• Biomarkers – many different roles, utilities– Screening; diagnosis; predictive; prognostic
• Genomic or proteomic (protein): uses– Eligibility – enrich the study population– Stratification– As study endpoint for efficacy or toxicity - NO
40
Tumor assessment endpoints (7)
• Biomarkers: not reliable to date as predictive of response outcomes– CEA failure to correlate for colon– CA-125 ovarian ?– PSA response failed to correlate with survival
Tax 327 approval study example
41
Random
ize
Mitoxantrone 12 mg/m2 Q3 weeksPrednisone 10 mg q dayup to 10 cycles
Docetaxel 75 mg/m2 Q3 weeksPrednisone 10 mg q dayup to 10 cycles
Docetaxel 30 mg/m2 Q 1 wkPrednisone 10 mg q day5 on; 1 off x 6 cycles
N=1006
Biomarker “surrogacy” from TAX 327 study
M Eisenberger, et al. Proc ASCO, 2004. Abs 4.
42M Eisenberger, et al. Proc ASCO, 2004. Abs 4; D Petrylak, et al. Proc ASCO, 2004. abs 3.
Docetaxel 3-weekly
Docetaxel weekly
M+P
PSA response
45 48* 32
Pain response
35% 31% 22%
Meas. Dis response
12% 8% 7%
Med OSmonths
18.9 17.4 16.5
Logrank p
HR (95% C.I.)
versus M+P
0.009
HR=0.76 (0.62, 0.94)
0.36 *
HR=0.91 (0.75, 1.11)
Docetaxel HRPC TAX 327 Trial- PSA endpoint
43
Tumor assessment endpoints and approval (7)
• RR / TTP / PFSHope – get results faster from studies
Often used in phase 2 studies with single arm
Hard to judge in single arm - no comparator
• Are these “surrogate” endpoints ?Problem - inconsistent relation to clinical benefit in
phase 3 trials
• Valid surrogates: Blood pressure / Cholesterol
– Strong and consistent relation with outcomes
44
Approval and tumor assessment endpoints (8)
Regular Approval – RA– If Clinical Benefit shown (live longer, better, safer) - or -– Benefit on Established surrogate - (DFS, Heme CR)
Accelerated Approval - AA For Serious or Life Threatening illnesses• Show meaningful therapeutic benefit over
existing therapy or improved patient response over available therapy
• May be based on a surrogate endpoint which is reasonably likely to predict clinical benefit - or
• a clinical endpoint other than survival or irreversible morbidity
45
Tumor assessment endpoints: TTP (9)
• Advantages:– Smaller sample size, shorter follow-up– Results sooner than a survival endpoint – Not affected by subsequent Rx– Determined by the entire treated group,
not just responders (= response duration)– May correlate with delay of new or more severe
symptoms or complications of malignancy
46
Tumor assessment Endpoints: TTP (10)
• Limitations– Most oncology trials unblinded =>
assessments may be subject to bias• Physicians may choose when to assess a particular
symptom or a tumor marker• Missing measurements of target / non-target lesions• Asymmetry of assessment timing
• Need precise prospective definition of TTP• Consistent supportive evidence also
– RR, PRO
47
TTP in Breast Cancer (11)
• ODAC Meeting 6-99 examined TTP in MBC• MBC using First-line cytotoxic treatment
• TTP not shown to correlate with OS or QOL– May be a problem of TTP magnitude, later Rx
• ODAC advised: TTP not be used alone for RA– Epirubicin: not approved for MBC – study used
primary endpoint TTP
• Potential use as surrogate endpoint for accelerated approval in RCTs with same study continued to survival endpoint
48
What about Response - CR or PR (12)
• CR = complete response PR = partial response– assessment must be prospectively defined: when, how– published criteria available to define response
• RECIST, NCI-WG, IBMT, etc
– In phase 2, response => ACTIVITY, not benefit– Is it associated with symptom benefit?– Durability of response is important
• MBC with hormone Rx: RR + TTP may => RA• Durable CRs in Hematologic malignancies
– fewer infections, visits, transfusions –
– may => RA
49
What about Response - CR or PR (13)
• Examples of CR based regular approvals:– Cladribine for hairy cell leukemia (CRs > 8mo)
– Pentostatin for hairy cell leukemia CRs > 24 mo
– Ifosfamide in combination for 3rd-line therapy of germ cell testicular tumors (CRs > 2 years)
– IL-2 for renal cell carcinoma / melanoma – cures ?
– AsO3 and ATRA for APL – RA: small, single arm
50
Approval and tumor assessment endpoints (14)
• AA uses tumor assessment endpoints Magnitude and consistency of effect importantRR + TTP may achieve AA TTP alone unlikely to be sufficientRR + TTP with hormone Rx – breast cancer may => RA CRs with Duration – Hematologic cancers may => RA
Endpoints and FDA Oncology Drug ApprovalsJohnson, Williams, PazdurJournal of Clinical Oncology 2003; 21:1404-1411
• Disease - Endpoint – Drug - Approval
51
Effectiveness: Oncology Trial Design (1)
How to show effectiveness – 1. Choose appropriate endpoint (with FDA)
• Disease, therapy, and regulatory context
2. Choose appropriate study design (with FDA)
– Phase 2 or phase 3 – many phase 2 results not confirmed in phase 3!
– Usually phase 3 with comparator arm of:• Standard of care treatment or Placebo
• Prospective, randomized, blinded * comparison– Blinding or masking: allocation, investigator, patient, sponsor
52
Effectiveness: Oncology Trial Design (2)
2. Appropriate study design: How to compare?• Type of comparison – Superiority vs. Non-inferiority
– Superiority: • versus placebo or add-on design • “head to head” with an active control Rx (risky)
– Non-inferiority – problems IN ONCOLOGY
• Estimating effect of control treatment• Constancy assumption - historical control• Retention margin• Large sample sizes required
53
Effectiveness: Oncology Trial Design (3)
3. Appropriate Analysis Plan– Pre-specified - why?
• Control error rate (chance of false positive conclusion)
• Not post-hoc, data-driven– Estimate the difference to be detected
• What can your new drug do?
– Size the study – what power do you want to have to demonstrate the primary endpoint
– ITT population for comparison of arms – Why ?
54
Effectiveness: Oncology Trial Design (4)
– ITT population for comparison of arms – Reduces Bias
• Cannot define the analysis population AFTER data examined
– ITT to see differences between arms
55
Effectiveness: Oncology Trial Design (5)
4. Statistical meaning– Null hypothesis = assume no difference between groups– Stat test – How likely is this difference a result of chance?– If unlikely due to chance, maybe due to treatment?– FDA concern - Error of a false positive conclusion (1/20)
• ? Acceptable error rate – pre-specified = Alpha • two-sided 0.05 - generates our P < 0.05
– If there’s no difference found? ≡ Equivalence?
• No! –Only means you cannot reject null hypothesis– Bayesian perspectives
4. Clinical meaning beyond statistical meaning
56
Oncology Approval Example - Prostate
• 2004 Docetaxel (Taxotere) approval for HRPC
• 2003 - Approved drugs for treatment of HRPC
– Estramustine 1981
– Mitoxantrone 1996 *
– Zoledronic acid 2003 *
* approvals based on QOL-PRO not on survival
57
Random
ize
Mitoxantrone 12 mg/m2 Q3 weeksPrednisone 10 mg q dayup to 10 cycles
Docetaxel 75 mg/m2 Q3 weeksPrednisone 10 mg q dayup to 10 cycles
Docetaxel 30 mg/m2 Q 1 wkPrednisone 10 mg q day5 on; 1 off x 6 cycles
N=1006
TAX 327
SWOG 9916
Random
ize
Mitoxantrone 12 mg/m2 Q 3 weeksPrednisone 5 mg bid
Docetaxel 60 mg/m2 d 2 Q 3 weeksEstramustine 280 mg d1-5*Dexamethasone 20 mg, tid d 1 & 2
N=770
*Warfarin and aspirin
M Eisenberger, et al. Proc ASCO, 2004. Abs 4; D Petrylak, et al. Proc ASCO, 2004. abs 3.
58
Docetaxel HRPC TrialsStatistical Designs
TAX 327• Power: 90% to detect a
25% reduction in hazard of death (HR=0.75)
• Accrual - 1006 patients
• Two studies available -
HRPC – hormone-refractory prostate cancer
SWOG 9916• Power: 80% to detect a
33% reduction in hazard of death (HR=0.67)
• Accrual - 770 patients
not just one
59
Mediansurvival Hazard
(mos) ratio P-value
D Combined: 18.2 0.83 0.03
D 3 wkly: 18.9 0.76 0.009D wkly: 17.3 0.91 0.3Mitoxantrone 16.4 – –
Months
Pro
bab
ilit
y o
f S
urv
ivin
g
0 6 12 18 24 30
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Docetaxel 3 wkly
Docetaxel wkly
Mitoxantrone
M Eisenberger, et al. Proc ASCO, 2004. Abs 4.
Survival (overall survival) -TAX 327
---------------------------------------------------
60
“Statistically Significant” Results
• Response rates – proportions - Chi square test • Survival curves – KM - time to event (OS, PFS)
– Median survival – midpoint on KM curves• Why not 75% or 25% ?• Why not 1 year or 2 year survival (% alive at __)
– P value - logrank test comparing the survival distributions (curves)
– Hazard ratio – Cox model assumptions– Confidence intervals (not overlap 1)
• Clinically significant ?
61
Statistical / Clinical Significance ?
• Best case – have both
• Can you have stat sig. but not clinical - yes– Approval likely ? No
• Can you have clinical sig. but not stat - yes– Approval likely ? It depends
• Phase 2 results• Statistical design may become inappropriate• Safety advantage, Other• Clinical benefit is the goal!
62
FDA Oncology Drug Approval
• Regulatory consequences of
Demonstrating Efficacy
• Regular Approval if: clinical benefit endpoint – OS, QOL, DFS, and occasional other
• Accelerated Approval if:• Benefit over existing therapy – if any; and• If a surrogate, then “reasonably likely” to predict, and• Must verify clinical benefit later
• If efficacy uncertain – ODAC likely
63
Evidence for Accelerated Approval
• Substantial evidence from well controlled clinical trials regarding a surrogate endpoint
• NOT: Borderline evidence regarding a clinical benefit endpoint
64
FDA Review Times
• Assuming a complete application and no substantive amendments submitted during the review
• Priority review completion: – 6 months– May fulfill an unmet medical need– Substantial improvement
• Standard review completion: – 10 months
65
Ages of Oncology Drug Approval
• Historical Era– Response rates - approval
• Current Era– Statistical refinements– Clinical Benefit and surrogate endpoints
• Molecular Era
66
Molecular Era (1)
• Characterizing both Disease and Patient– Individual Patient characteristics
• Individualized dosing – PG profile of each person including CYPs, receptors, transporters
– Individual Tumor characteristics• Receptors, transporters• Targets
– Enriched populations with target– Study designs to assess role of the target
• More selective drug effects
• Individualized therapy
67
Molecular Era (2)
• Phenotype to Genotype: will redefine the disease, the patient, and the indication
• Phenotype to Genotype - examples– MCL: Cyclin D1 over-expressed or t(11,14) CD20
B cell lymphoma of nodes, spleen and marrow– NSCLC: “Taressa” indicated for EGFR Exon 20 activating
mutation, Bcl-2 overexpressed, cancer in lung or kidney except for CYP2D6 slow metabolizers
• Hope – we no longer have to tell someone –
there’s a chance this therapy may help you – but we can’t predict …
68
Oncology targeted therapies 2006 (3)
Monoclonal Antibodies: approval target label ?Rituxan (Rituximab) – lymphoma 1997 CD20 yesHerceptin (Trastuzumab) - breast 1998 p185neu yes*Mylotarg (Gemtuzumab) - AML 2000 CD33 yes*Campath (Alemtuzumab) – CLL 2001 CD52 no *Erbitux (Cetuximab) – colon 2004 EGFR yesAvastin (Bevacizumab) – colon 2004 VEGF no
Label ? - Is a test for the target included in the label indication?(May not be exactly the same test as the target assay)
(Radio-immunoconjugates- Zevalin (2002) and Bexxar (2003): anti-CD20)
*AA products - Comparative, randomized trials demonstrating increased survival or clinical benefits such as improvement in disease-related symptoms have not yet been conducted.
69
Oncology targeted therapies 2006 (4)
• Small molecule inhibitors approval target label ?
– Nolvadex (Tamoxifen)-Breast 1977 ER yes
– Vesanoid (ATRA) ----- APL 1995 RARα yes
– Gleevec (Imatinib) ---- CML 2001 bcr/abl yes
– *Gleevec (Imatinib) -- GIST 2002 c-kit yes
– *Iressa (Gefitinib) ----- Lung 2003 EGFR? no
– Tarceva (Erlotinib) ---- Lung 2004 EGFR? no
Label ? – Is a test for the target included in the label indication?
*AA -comparative, randomized trials demonstrating increased survival or clinical benefits such as improvement in disease-related symptoms have not yet been conducted – AA
70
Allsubjects
Responders
Drug treatment
Non-responders
Usual target study plan - Retrospective
Target- subjects
Target+ subjects
Target- subjects
Target+ subjects
Problem: retrospective and subgroup analysis- Imbalances
71
Allsubjects
Target POS subjects
All Target-tested
Target NEG subjects
Prospective, Stratified: assess effect in Target POS and NEG patients
Control
Drug
Control
Drug
For target assay utility where predictive value unknown or well below 100% – focus is on the target
®
®
® = randomization
72
Prospective, Stratified Evaluation of Biomarker
All subjects All PG tested
Test is -
Test is +Placebo
Drug
Placebo
Drug
All subjects
All PG tested but not available at randomization Placebo
Drug Analyze PG+ as primary analysis
Analyze toxicity in all
Possible effect in the (-) group and/or toxicity in the (-) group needs to be evaluated because pre-treatment selection is not possible
73
Targeted Therapy - Problems• Which target(s) is the target?
Herceptin target-gene amplification not protein “expression”
• Can’t measure the target – EGFR, her-2
• Can’t correlate target inhib. with outcome– Wrong target for the disease state– Target may not be in the disease pathway– Variable “expression” “over-expression”– Target present but non-functional or variable f(x)
• Can’t validate target – “Inconclusive” study design
74
Cell Cycle Regulatory Pathways
75
FDA Oncology Drug Approval
• Appropriate Endpoint• Appropriate Design• Appropriate Conduct (FDA will verify data)• Appropriate Analysis• Demonstrate Efficacy and Safety
(Benefit - Risk assessment)DDOP - NDA => Marketing Approval
DBOP - BLA => Biologics license
76
General NDA Review Procedure for a New Drug
• Separate reviews by disciplines – stat, med, pharm-tox, Biopharm, CMC
• All primary data reviewed
• Analysis of Benefit versus Risk in the context of the disease process.
• Applications may be discussed before an advisory committee (ODAC) at an open public meeting
77
Why Bad Things (non-approval)can happen to Good Drugs
78
“Tortured Data Will Eventually Confess”
Some Examples of Clinical Trial Conduct “problems”– Inadequate or no controls– Missing or unclear selection/eligibility criteria– Small sample size - underpowered– Randomization process concerns– Lack of objective outcome assessment– Improper handling of dropouts– Inadequate adjustment for prognostic factors– Improper or misleading tables and graphs
79
“Complementary and Alternative Statistics”
• Some improper statistical methods– Disregard of multiple comparisons– Improper (selective) censoring, exclusions– Post-hoc hypothesis selection (data-dredging) usually
on a subgroup analysis not pre-specified– Claiming subgroup results when the overall study fails
80
notes• ODAC 1977:
– Approval to be based on survival or, possibly, improved quality of life
• Supreme Court 1979: US vs. Rutherford (Laetrile) – A drug is effective “if it … prolonged life,
improved physical condition, or reduced pain”• ODAC March 24, 1983:
– Reaffirmed above, and added “objective tumor response could also be used if a positive correlation between tumor response and: survival, QOL, or relief of pain could be shown.”
81
FDA Website for Oncology Endpoints
www.fda.gov/cder/drug/cancer_endpoints/default.htm
FDA Guidances:
www.fda.gov/cder/guidance/index.htm
• Guidance on Effectiveness:http://www.fda.gov/cder/guidance/5506fnl.pdf
• Guidance on Oncology trials endpoints- draft
http://www.fda.gov/cder/guidance/6592dft.htm • Guidance on PRO - draft
http://www.fda.gov/cder/guidance/5460dft.htm
82
23. This isn't an office. It's Hell with fluorescent lighting.
24. I started out with nothing & still have most of it left.
25. Sarcasm is just one more service we offer.
26. If I throw a stick, will you leave?
27. Errors have been made. Others will be blamed.
28. Whatever kind of look you were going for, you missed.
29. I'm trying to imagine you with a personality.
30. A cubicle is just a padded cell without a door.
31. Can I trade this job for what's behind door #1?
32. Too many freaks, not enough circuses.
33. Nice perfume. Must you marinate in it?
34. Chaos, panic, & disorder-- my work here is done.
35. How do I set a laser printer to stun?
36. I thought I wanted a career; turns out I just wanted a salary.
37. Oh I get it ... like humor ... but different
83
11. I like you. You remind me of myself . . . when I was young andstupid.
12. You are validating my inherent mistrust of strangers.
13. I have plenty of talent and vision; I just don't give a damn.
14. I'm already visualizing the duct tape over your mouth.
15. I will always cherish the initial misconceptions I had about you.
16. Thank you. We're all refreshed and challenged by your unique pointof view.
17. The fact that no one understands you doesn't mean you're an artist.
18. Any connection between your reality and mine is purely coincidental.
19. What am I? Flypaper for freaks!?
20. I'm not being rude. You're just insignificant.
21. And your crybaby whiny-assed opinion would be...?
22. Do I look like a people person?
84
Notes:Basis for New Drug Approval - NDA
• Demonstration of efficacy with acceptable safety in adequate and well-controlled studies CFR 314 - NDA Regs
• Ability to generate product labeling that– Defines an appropriate patient population for
treatment with the drug– Provides adequate information to enable safe
and effective use – prescribing of the drug
• Analogous rules for Biologics - BLA
85
United States versus RutherfordUS Supreme court 442 US 544 (1979)
• A drug is effective “if it fulfills by objective indices, its sponsor’s claims of prolonged life, improved physical condition, or reduced pain,” and safe if the drug’s potential for inflicting death or physical injury is offset by the possibility of therapeutic benefit.
86
notes• ODAC 1977:
– Approval to be based on survival or, possibly, improved quality of life
• Supreme Court 1979: US vs. Rutherford (Laetrile) – to be effective, a cancer drug must by objective indices,
improve survival, improve the quality of life, or relieve pain• ODAC March 24, 1983:
– Reaffirmed above, and added “objective tumor response could also be used if a positive correlation between tumor response and: survival, QOL, or relief of pain could be shown.”
• ODAC June 28, 1985: – Prolonged disease-free survival is an important goal
of adjuvant studies and is sufficient for approval of a drug for adjuvant therapy of breast cancer
87
Notes
• FDA Modernization Act 1997: FDAMA– Created CFR 312 subpart E: Procedures to expedite
development for serious and life-threatening disease– Fast track – process for meeting with FDA– Priority review – 6 month NDA review time frame for
products addressing unmet medical need– Endorsed as possible one high quality study
• SPA – clinical protocols for phase 3 studies forming primary basis for efficacy for NDA; 45 d.
88
FDAMA 1997 – one study for efficacy• 1998 FDA Guidance:
Characteristics of a single study to support effectiveness (with independent substantiation from related study data):a. Large, multicenter study
no single site or investigator disproportionately responsible for resultb. Consistency across study subsets
consistency across key subsets, i.e. severity of disease, stage, age
c. Multiple studies within the study – pairwise comparisons within the study
d. Multiple endpoints involving different eventssomewhat unrelated endpts i.e. MI and death
e. Statistically very persuasive – very low p values (not 0.045!)
89
U.S. Legal Process
• Regulations: CFRs– Interpretations of laws – by the Executive Branch Departments – FDA Regs: Full power of laws when adopted
• Guidance: Issued by individual agencies (FDA or CDER) to reflect current thinking, not binding.
90
Chemo-Prevention Endpoints
Current Approvals
91
Reduction in cancer incidence
• Tamoxifen is approved:– to reduce the incidence of BC in women at
high risk for BC– In women with DCIS, following breast surgery
and RT, to reduce the risk of invasive BC– To reduce contralateral BC in patients
receiving adjuvant tamoxifen therapy for BC• Basis: Reduction in cancer incidence in RCT
92
Reduction in polyp number
• Celecoxib approved to reduce the number of adenomatous colorectal polyps in FAP as an adjunct to usual care
• Basis:
• Reduction in polyp number
• Accelerated approval - RCT